research article leaf proteome analysis reveals...

Research ArticleLeaf Proteome Analysis Reveals Prospective Drought andHeat Stress Response Mechanisms in Soybean

Aayudh Das,1,2 Moustafa Eldakak,1 Bimal Paudel,1 Dea-Wook Kim,3 Homa Hemmati,4

Chhandak Basu,4 and Jai S. Rohila1

1Department of Biology & Microbiology, South Dakota State University, Brookings, SD 57007, USA2Department of Biochemistry and Biophysics, Texas A&M University, College Station, TX 77840, USA3National Institute of Crop Science, Rural Development Administration (RDA), Wanju-gun, Jeollabuk-do 55365, Republic of Korea4Department of Biology, California State University Northridge, Northridge, CA 91330, USA

Correspondence should be addressed to Jai S. Rohila; [email protected]

Received 10 November 2015; Accepted 1 February 2016

Academic Editor: Paloma K. Menguer

Copyright © 2016 Aayudh Das et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Drought and heat are among the major abiotic stresses that affect soybean crops worldwide. During the current investigation, theeffect of drought, heat, and drought plus heat stresses was compared in the leaves of two soybean varieties, Surge and Davison,combining 2D-DIGE proteomic data with physiology and biochemical analyses. We demonstrated how 25 differentially expressedphotosynthesis-related proteins affect RuBisCO regulation, electron transport, Calvin cycle, and carbon fixation during drought andheat stress. We also observed higher abundance of heat stress-induced EF-Tu protein in Surge. It is possible that EF-Tu might haveactivated heat tolerance mechanisms in the soybean. Higher level expressions of heat shock-related protein seem to be regulatingthe heat tolerance mechanisms. This study identifies the differential expression of various abiotic stress-responsive proteins thatregulate various molecular processes and signaling cascades. One inevitable outcome from the biochemical and proteomics assaysof this study is that increase of ROS levels during drought stress does not show significant changes at the phenotypic level inDavisonand this seems to be due to a higher amount of carbonic anhydrase accumulation in the cell which aids the cell to become moreresistant to cytotoxic concentrations of H

2O2.

1. Introduction

Soybeans are one of the most important legume crops andhave a major impact on the US and global economies. Inter-national markets use about half of the soybeans producedin the US, and its production is expected to alleviate theglobal demand for human consumption, biofuel production,and high-protein meal for animal feed [1]. In 2011, the totalsoybean production in the US was 83.29 × 1012 Kg with atotal trade value of $40.2 billion [2, 3]. Heat and droughtare the predominant abiotic stress factors that limit thegrowth and development of soybean plants by causing areduction in carbon fixation by the photosynthetic apparatusof the plants, resulting in net yield losses [4]. At the cellularlevel, plants exhibit a variety of responses related to theirphysiology and biochemistry to overcome the stress. Droughtstress causes reduced carbon assimilation due to stomatal

closure, membrane damage, and distressed activity of variousCO2fixation enzymes [5]. Heat stress increases membrane

damage and impairs metabolic functions [6–8]. As a result,combination of drought and heat stress causes enormouseconomic losses for farmers [9].

Proteomic evaluation for the purpose of differentiallyregulated proteins identification in response to the drought,heat, and the combined stress has been monitored in variousplants in context of the morphological and physiologicalchanges [10–12]. Particularly, changes in proteomic expres-sion during drought stress have been observed in majorcrops like rice and wheat which shows differential regulationof various proteins [13–15]. Earlier studies using proteomicand transcriptomic approaches also implicated the droughtstress response mechanisms including alteration in the sig-nal transduction pathway and plant’s metabolism [16–19].Moreover, in response to heat stress, many stress proteins

Hindawi Publishing CorporationBioMed Research InternationalVolume 2016, Article ID 6021047, 23 pageshttp://dx.doi.org/10.1155/2016/6021047

2 BioMed Research International

were originally identified as HSP which indicates plant’stolerance threshold [20]. To improve the understanding ofthe mechanisms underlying soybean responses to droughtand heat stress, in the present investigation we have useda global proteomic approach combined with physiologicaland computational analysis [10, 21, 22]. The two majorsoybean varieties, Surge (yield 3,272Kg/ha) and Davison(yield 4,074Kg/ha), showing distinct physiological charac-teristics were used in this study. Proteomic analysis revealeddifferentially expressed proteins related to key biologicalprocesses (such as photosynthesis and respiration), variousmetabolic pathways (nitrogen metabolism and carbohy-drate metabolism), and several other molecular processes(protein biosynthesis and ATP synthesis). Computationalanalysis predicted a protein network displaying the likelyinteractions between protein abundances and plant stressresponses. These analyses will enable the development ofplant molecular improvement programs to produce moreeffective strategies contributing to greater food security in thecoming years. Since the proteins are the translated versionof mRNA (proteomics approach may have advantage overtranscriptomics approach [4]) several research groups haveutilized the power of proteomic evaluations and combinedit with physiological studies to create a link between plantphysiology and the molecular signatory responses [10, 23].The primary aim of this study was to investigate differentialprotein expression between two contrasting genotypes fordrought and heat stress responses, and to reveal the droughtand heat stress-responsive mechanisms in soybean at earlystage of growth. Based on physiological analyses, we foundthat drought and heat stress decrease photosynthesis andreduce stomatal conductance and transpiration rates, whichultimately alter the net CO

2concentrations in leaves. The

proteomic analyses facilitated the characterization of a setof proteins whose expressions were altered under heat anddrought stresses.We found that, in Surge, more proteins weredownregulated during drought stress than during heat stress,but the combined stress conditions exerted a drastic effect atboth the molecular and the phenotypic levels. In this study,we have identified genes involved in photosynthesis thatwere differentially expressed during drought and heat stressconditions. This differential expression is most likely theresult of blocked interactions between RuBisCO, RuBisCOactivase, electron transport chain, and downregulation ofvarious Calvin cycle enzymes, which ultimately results in anet reduced level of photosynthesis [24].This study identifiedgreater accumulations of EF-Tu protein during heat stress inSurge.The results also indicated the accumulation of a 70 kDastromal HSP, which we assume is crucial for plant survivalduring heat stress by activating heat tolerance responses [25].Previously, it has been reported that drought and heat stressesmay trigger the formation of reactive oxygen species (ROS)[26, 27]. Likewise, we also examined the drought and heatstress-induced changes in the production of ROS and foundthat, during drought stress, soybeans exhibited a significantincrease in the level of hydrogen peroxide and an enhancedROS detoxification capacity via carbonic anhydrase, whichprotects the plant against oxidative damage [28]. Plants useABA as a signaling molecule while maintaining a positive

water balance throughout the system, as physiological studieshave shown [29–31]. Elevated ABA levels during heat stressare shown to protect the maize plant against heat-inducedoxidative damage [32]. We quantified the ABA levels insoybean leaves during all the stress conditions and have dis-cussed it in drought and heat stress contexts for soybean crop.At the level of computational analysis, we have constructed aprotein-protein network to predict the interactions betweenthe differentially expressed proteins [17].

2. Materials and Methods

2.1. Plant Growth Conditions. Soybean (Glycine max L.cultivars: Surge and Davison) seeds were planted in potsfilled withMetromix 360 (Sun GroHorticulture, Floodwood,Minnesota, USA). A total of eight, 6.0-liter size pots, eachcontaining 4-5 individual soybean plants (biological replica-tions), were placed in a growth chamber (Conviron, Canada)illuminated with white fluorescent light and incandescentbulbs (500 𝜇molesm−2 s−1, 12 h photoperiod) at 25∘C and70% RH for 3 weeks. At this stage, 8 pots from each cultivarwere separated into 4 groups (control, drought, heat, anddrought plus heat; Figure S1 in Supplementary Materialavailable online at http://dx.doi.org/10.1155/2016/6021047).Watering was done daily with 150mL tap water per pot untilthe second trifoliate leaves emerged, afterwhich various stresstreatments were carried out. The first group was the control,and it was maintained under the conditions described above.For the second group, the plants were exposed to droughtstress (no watering for 7 days). The third group containedplants that were exposed to heat stress (42∘C daytime/35∘Cnight time, 50% humidity) with normal watering schedules.The fourth group contained plants that were exposed todrought plus heat stress. Soil moisture sensors (SM 100,Spectrum Technologies) were inserted into all of the pots onthe day the plants were exposed to abiotic stress to measurethe % VWC for each treatment. The % VWC value wasmeasured throughout the entire experimental period usinga microstation data logger [33].

2.2. Measurement of Photosynthesis, Stomatal Conductance,Transpiration, and Leaf Water Potential. Photosynthesis,stomatal conductance, and the transpiration rate were mea-sured, every day at noon, on the first and second trifoliatestage leaves for all the plants in each group and with threereplications (following the rationale from previous publisheddrought stress time point studies [34, 35]). We used CI-510Photosynthesis System (CID Bio-Science) with the followingparameters: leaf chamber area: 6.25 cm2; flow rate: 0.3; timeinterval: 1 second; delay: 2 seconds; system: open; tempera-ture: 25∘C, as per the manufacturer’s recommendations. Leafwater potential wasmeasured using a pressure chamber (PMSInstrument) following the manufacturer’s recommendations.

2.3. Sample Preparation for 2D-DIGE. Leaf samples collectedfrom all treatments were snap-frozen under liquid nitrogenand stored in −80∘C freezer until further use. Total pro-tein was isolated according to the procedure of Hurkman

BioMed Research International 3

and Tanaka [36]. Ground soybean leaf tissue (500mg)was homogenized in 5mL of 2D cell lysis buffer (30mMTris-HCl, pH 8.8, 0.9M sucrose, 10mM EDTA, 0.4% 2-mercaptoethanol, 7M urea, 2M thiourea, and 4% CHAPS)and an equal volume of Tris-saturated phenol, vortexedfor 30 seconds and incubated for 30min at 4∘C. Followinghomogenization and mixing, centrifugation was carried outfor 15min at 4∘C at 6,000 rpm, and the phenol phase wascollected in a fresh tube. Now, the proteins in the collectedphase were precipitated overnight by adding 5 volumes of ice-cold 0.1M ammonium acetate in 100% methanol at −20∘C.Next day, after 20min of centrifugation at 10,000 rpm, theprotein pelletwaswashed in 5mLof 0.1Mammoniumacetatein 100% methanol followed by a wash in 5mL of ice-cold80% acetone, and then a final wash in 4mL of 70% ethanol.The protein pellet was air-dried and stored at −80∘C untildownstream experiments. Protein quantification assays wereperformed using the Bio-Rad reagent (catalog# 500-0006) onthe Bio-Rad SmartSpec Plus spectrophotometer.

2.4. 2D-DIGE, Trypsin Digestion, and MALDI-TOFMS Anal-ysis. 2D-DIGE analysis was performed by Applied Biomics(Hayward, CA, USA). For each sample set (either Surgeor Davison with respective treatments, Figure S1), 30 𝜇g ofsample protein was mixed with 1.0 𝜇L of diluted Cy3 or Cy5dye (1 : 5 dilution in a 1 nmol𝜇L−1 DMF stock) for labeling.A pooled protein sample containing equal amounts of allsamples in the experiment was labeled with Cy2 using thesame protocol. After vortexing, the tubes were incubated inthe dark for 30min on ice. Next, 1.0 𝜇L of 10mM lysine wasadded to each sample. After that, the samples were vortexedand incubated in the dark for 15min on ice. Next, the Cy2,Cy3, and Cy5 labeled samples were mixed by adding 2X 2Dsample buffer (Bio-Rad), followed by the addition of 100 𝜇Lof DeStreak solution and a rehydration buffer (Bio-Rad) upto 350 𝜇L for the 13 cm gradient (pH 4–7) IPG strip. Uponcompletion of IEF, the IPG strips were incubated in freshlymade equilibration buffers I and II (Bio-Rad) for 15 minuteswith gentle shaking. The IPG strips were rinsed in Tris-glycine-SDS running buffer (Bio-Rad) and then transferredto 12% SDS-polyacrylamide gels, followed by sealing with0.5% agarose solution for second-dimension electrophoresis.The SDS gels were electrophoresed at 175 volts at 15∘Cfor 8 hrs. Gel images were scanned using Typhoon TRIO(GE Healthcare, PA, USA) and were analyzed using ImageQuantTL software (version 6.0, GE Healthcare, USA) andthen subjected to in-gel analysis and cross-gel analysis usingDeCyder software, version 6.5 (GE Healthcare). The dif-ferential expression of the proteins was obtained from in-gel DeCyder software analysis [37]. This way, a total of 12gels (Table S1) were run comprising protein samples from 8different sets of treatments (Figure S1) and 3 replications each.The spots of interest were selected based on the in-gel analysisand statistical analyses (Table S2), 𝑝 value of ≤0.1 and cut-off value of 1.5-fold, and were picked up using the Ettan SpotPicker (GE Healthcare). The picked gel spots were washedand digested in-gel using modified porcine trypsin protease(Trypsin Gold, Promega). The digested tryptic peptides weredesalted using Zip-tip C18 (Millipore) and spotted on the

MALDI plate. MALDI-TOF MS and TOF/TOF tandem MSwere performed on a 5800mass spectrometer (AB Sciex).TheMALDI-TOF mass spectra were generated in the reflectronpositive ion mode, and TOF/TOF tandemMS fragmentationspectra were acquired for each sample. On average, 4,000laser shots per fragmentation spectrum were applied to eachof the 5–10 most abundant ions present in each sample,excluding the trypsin autolytic peptides and other knownbackground ions [38]. Both the resulting peptide mass andthe associated fragmentation spectra were submitted to theMASCOT (version 2.4,Matrix Sciences, UK) inOctober 2014to search the SwissProt database with 546,790 sequences.Thesearches were performed following the details of Poschmannet al. [39], for example, without constraining the proteinMWor the pI value and with variable carbamidomethylationof cysteine and oxidation of methionine residues, a masstolerance of 100 ppm, and allowing only one missed and/ornonspecific cleavage. Proteins with a protein score > 62 wereconsidered to be assigned correctly. Candidates with eithera protein score of CI % or an ion score of CI % greaterthan 95 were considered to be statistically significant. Theprotein peptide summary for all identified spots is listed inSupplementary Material Table S4.

2.5. ROS Measurement (Hydrogen Peroxide Quantification).The levels of H

2O2in leaves were measured using the

Amplex� Red Hydrogen Peroxide/Peroxidase Assay Kit(Molecular Probes, catalog# A-22188) [40]. For this assay,manufacturer’s protocol was followed. Briefly, 50mg of theground leaf powder was mixed with reaction buffer (pH7.4), vortexed, and incubated on ice for 15min. Then, thesamples were centrifuged at 10,000 rpm for 10min at 4∘C,and the supernatant was collected. Next, 50 𝜇L of the sampleswas loaded into a multiwell plate (clear bottom, black sides)followed by the addition of 50 𝜇L of 100 𝜇M Amplex� Redreagent and 0.2UmL−1 HRP. The samples were incubatedin the dark for 30min at room temperature. Afterward,the fluorescence was measured at an emission wavelengthof 620 nm using a Synergy 2 multimode microplate reader(BioTek); the blank value (buffer only) was subtracted fromthe sample’s florescence values. Then, based on the standardcurve, the H

2O2concentration was calculated in terms of

nmol𝜇L−1 sample [41].

2.6. Quantification of Abscisic Acid. ABA quantification wasperformed as outlined by Kim et al. [42]. Two hundredmg of ground leaf powder was suspended in 1mL of steriledeionizedwater and incubated overnight at 4∘Cwith constantshaking. The solution was centrifuged at 4,000 rpm for20min. Then, the supernatant was transferred to a cleanmicrofuge tube ensuring that no leaf pieces were transferred.The solution then was dried completely using a vacuum con-centrator. The dried precipitate was resuspended with 60𝜇Lsterile deionized water and a 1 : 1000 dilution of the samplewas made with TBS. ABA concentration was determinedusing a Phytodetek� ABA Test Kit (Agdia, USA) followingthe manufacturer’s recommendations. A standard curve wasgenerated using ABA standards (100, 20, 4, 0.8, 0.16, and0.32 pmolesmL−1). 100𝜇L of diluted sample per standardwas


Control Drought Heat Drought + heat0.00

−0.50

−1.00

−1.50

−2.00

−2.50

−3.00

−3.50

−4.00

−4.50

Leaf

wat

er p

oten

tial (

MPa

)

Different abiotic stress treatment

−1.63

−2.70

∗∗∗

∗∗∗∗∗∗

−2.35

−3.83

∗∗∗p < 0.01

(a)

Leaf

wat

er p

oten

tial (

MPa

)

Different abiotic stress treatment

0.00

−0.50

−1.00

−1.50

−2.00

−2.50

−3.00

−3.50

−4.00

−4.50

−1.82

−3.60

−2.65

−3.80

∗∗∗∗∗∗

∗∗∗

Control Drought Heat Drought + heat

∗∗∗p < 0.01

(b)

Figure 1: Bar diagrams showing the effect of various abiotic stresses on leaf water potential in two soybean cultivars Surge (a) andDavison (b).𝑥-axis defines various abiotic stress treatments (drought, heat, and drought + heat) including control and 𝑦-axis defines leaf water potentialin MPa (megapascal unit). Each value represents the mean ± SE of three replicates and the asterisks designate the significance of changesfrom their corresponding control (∗∗∗𝑝 < 0.01).

loaded onto an ABA-antibody coated multiwell plate, andthe ABA concentration was determined according to manu-facturer’s guidelines. The absorbance values at 405 nm wereread using Spectromax M3 (Molecular Devices). The ABAconcentration was calculated using the slope of the standardcurve based on the manufacturer’s recommendations.

2.7. Computational Analysis and Protein-Protein InteractionPredictions. First, we mapped protein ID (using UniProtdatabase) to the Arabidopsis accession numbers and per-formed a multiple sequence alignment of Arabidopsishomologs usingCLUSTALO [43] to obtain the protein identi-ties. To search the variety of functional connections betweenthe 44 differentially expressed proteins, we used the onlinedatabase resource STRING (http://string-db.org/) version9.1 [44]. Then, after obtaining the primary interactions inSTRING, we portrayed our different protein networks usingCytoscape software, version 3.0.1 [45].

2.8. Clustering and Statistical Analyses. For the clusteringdata, the log

2-transformed expression values of the protein

spots were used. Hierarchical clustering of the proteinswas performed using Gene Cluster 3.0 [46] with Euclideandistance similarity metrics and the complete linkage method.The clusters were visualized using JAVA TREEVIEW [47].The statistical significance of the results was evaluated usingStudent’s 𝑡-test and a level of significance of 𝑝 ≤ 0.05for the two group comparisons. Data analyses and graph-ical representations were performed using Microsoft Excel2013.

3. Results

Understanding the mechanisms by which plants respond todrought, heat, and cooccurring drought and heat stressesplays a major role in optimizing crop performance underdrought and high temperature conditions [16]. Figure S2

shows the phenotypic changes between soybean plants on thesixth day of the stress experiment.

3.1. Soil Moisture Content and Physiological Responses ofPlants to Different Abiotic Stresses. To keep track of thesoil moisture levels during the stress experiments, the soilwater content was measured [48]. Figure S3 shows the soilmoisture levels from Day 0 (prestress) to the sixth day ofstress treatment. The leaf water potential is considered asa reliable parameter for quantifying the plant water stressresponse [49]. It was found (Figure 1) that, in Surge, stressreduced the leaf water potential from −1.63MPa (controlplants) to −2.70MPa in drought-stressed plants, −2.35MPain heat-stressed plants, and −3.83MPa in drought plus heat-stressed plants on the sixth day of stress. InDavison, we foundthat stress reduced the leaf water potential from −1.82MPa(control plants) to −3.60MPa in drought-stressed plants,−2.65MPa in heat-stressed plants, and −3.80MPa in droughtplus heat-stressed plants, on the sixth day of stress.

Photosynthesis is among the primary processes affectedby drought and heat stresses [50, 51]. In Surge control plants(Figure 2), it was found that there was no decrease inphotosynthesis, whereas in drought-stressed plants a 19%decrease in photosynthesis was detected on the sixth day ofstress. Interestingly, in heat-stressed plants, we found that thelevel of photosynthesis was identical to prestress condition;and in drought plus heat-stressed plants, we detected a27.28% decrease in photosynthesis on the sixth day of stresscompared to prestress (Day 0) conditions. In the Davisoncontrol plants, there was a 9% decrease in photosynthesismost likely due to a developmental effect and the water useefficiency of the plant on that particular day; in drought-stressed plants, there was a 34.36% decrease in photosyn-thesis; in heat-stressed plants, photosynthesis decreased to6.55%; and in drought plus heat-stressed plants, we detecteda huge decrease (28.13%) in photosynthesis on the sixth dayof stress compared to prestress (Day 0) conditions.


43.94

62.0253.68

49.7644.84

50.24 53.33

36.19

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

Different treatments

Prestress6th day of stress

SurgePre-stress 6th day of stress Difference Photosynthesis

Control 43.94 44.84 No dropDrought 62.02 50.24 11.79 19.00Heat 53.68 53.33 0.35 0.64

49.76 36.19 13.57 27.28Drought + heat

HeatDroughtControl

∗∗

∗∗

∗∗∗p < 0.01∗∗p < 0.03

Phot

osyn

thes

is (𝜇

mol

/m2/s

)

−0.90

(𝜇mol/m2/s) (𝜇mol/m2/s) (𝜇mol/m2/s) drop (%)

Drought + heat

(a)

63.11 63.86

52.0359.2857.25

41.9248.62

42.60

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

Prestress6th day of stress

DavisonPrestress 6th day of stress Difference Photosynthesis

drop (%) Control 63.11 57.25 5.86 9.28Drought 63.86 41.92 21.94 34.36Heat 52.03 48.62 3.41 6.55

59.28 42.60 16.67 28.13

Phot

osyn

thes

is (𝜇

mol

/m2/s

)

Different treatmentsDrought + heatHeatDroughtControl

∗∗∗∗∗

∗∗∗p < 0.01∗∗p < 0.03

Drought + heat

(𝜇mol/m2/s) (𝜇mol/m2/s) (𝜇mol/m2/s)

(b)

Figure 2: Bar diagram showing the effect of various abiotic stresses on photosynthesis in two soybean cultivars Surge (a) and Davison (b).𝑥-axis defines various abiotic stress treatments (drought, heat, and drought + heat) including control and 𝑦-axis defines measurement of netphotosynthesis (𝜇mol/m2/s). Each value represents the mean ± SE of three replicates and the asterisks designate the significance of changesfrom their resultant control (∗∗∗𝑝 < 0.01, ∗∗𝑝 < 0.03).

Several studies have suggested changes in the transpira-tion rate or stomatal conductance in response to differentstress conditions. Altering stomatal conductance causes afluctuation in the leaf water potential by shifting the tran-spiration rate [52]. The results (Figure 3) revealed that, inSurge control plants, stomatal conductance was increased by36.44% (at the same time, the transpiration rate increasedby 59.84%), whereas in drought-stressed plants stomatalconductance was decreased by 21.18% (while the transpira-tion rate decreased by 73.02%) on the sixth day of stresscompared to prestress (Day 0) conditions. In heat-stressedplants, stomatal conductance decreased by 15.94% (the tran-spiration rate decreased by 77.48%); and in drought plus heat-stressed plants, stomatal conductance decreased by 41.81%(the transpiration rate decreased by 130.35%) on the sixth dayof stress compared to prestress (Day 0) conditions. InDavisoncontrol plants, we found that stomatal conductance decreasedby 3.62% (the transpiration rate increased by 96.13%); indrought-stressed plants, stomatal conductance decreased by13.05% (the transpiration rate decreased by 77.37%); in heat-stressed plants, stomatal conductance decreased by 1.53%(the transpiration rate decreased by 14.06%); and in cooccur-ring drought plus heat-stressed plants, stomatal conductancedecreased by 15.45% (the transpiration rate decreased by133.43%) on the sixth day of stress compared to prestress (Day0) conditions.

Stomatal conductance also depends on the leaf tem-perature via the transpiration rates [52]. The analysis

of present investigation indicated that leaf temperatureincreased in a nonproportional manner with stomatal con-ductance (Figure 4). We found that stomatal conductancereduced under all conditions in which the leaf temperatureincreased.

3.2. 2D-DIGE Analysis Followed by MALDI-TOF MS toIdentify the Differentially Expressed Proteins in Response toDrought and Heat Stress. The comparison of the leaf proteinsfrom the two soybean cultivars among the control, heat,drought, and drought plus heat stress conditions via 2D-DIGE analysis revealed a broad distribution in the pI rangeand in themolecular weight range. Figure 5 is a representativeimage ofmaster gel, and supplemental Figures S4 and S5 showthe Cy2/Cy3/Cy5 overlay images of all 12 gels run duringthe experiment. The protein spots were selected from thepreparative gels for protein identification. Among the controland three treated groups, based on in-gel analysis, on averagea total of 2600 spots were detected using DeCyder software.Of these, on average a total of 1,900–2,000 protein spots werematched among all other gels (Table SI); from here, a totalof 108 spots were selected based on statistical analyses (𝑝values; Table S2), and 92 differential spots were successfullyidentified using a threshold of significance of 𝑝 ≤ 0.1, andwith a cut-off value of 1.5-fold increase/decrease in proteinexpressions. Of these identified spots, we determined theprotein identity of 88 spots with high confidence (Tables S3and S4), corresponding to 44 nonredundant differentially


36.44


50.00

40.00

30.00

20.00

10.00

0.00

−10.00

−20.00

−30.00

−40.00

−50.00

Chan

ge o

f sto

mat

al co

nduc

tanc

eat

mm

ol le

vel (

%)

HeatDroughtControl

−21.18−15.94

−41.81

Drought + heat

(a)Different treatments

0.00

−2.00

−4.00

−6.00

−8.00

−10.00

−12.00

−14.00

−16.00

−18.00

HeatDroughtControl

Chan

ge o

f sto

mat

al co

nduc

tanc

e at

mm

ol le

vel (

%) −3.62

−13.05

−15.45

−1.53

Drought + heat

(b)

59.84

Control Drought Heat


100.00

50.00

0.00

−50.00

−100.00

−150.00

−73.02 −77.48

−130.35

Tran

spira

tion

rate

chan

ge at

mm

ol le

vel (

%)

Drought + heat

(c)

96.13



−77.37

−14.06

−133.43

100.00

150.00

50.00

0.00

−50.00

−100.00

−150.00

−200.00

Tran

spira

tion

rate

chan

ge

at m

mol

leve

l (%

)

(d)

Figure 3: Bar diagram showing the comparison of stomatal conductance and transpiration rate profiles between Day 0 (prestress) and Day 6under various stress conditions in two soybean cultivars Surge ((a), (c)) andDavison ((b), (d)). 𝑥-axis defines various abiotic stress treatments(drought, heat, and drought + heat) including control and𝑦-axis defines percentage change of stomatal conductance atmmol level. Each valuerepresents the mean ± SE of three replicates.

20.00 20.96 21.39 19.9119.40 22.33 23.24 24.41

0.005.00

10.0015.0020.0025.0030.00

Control Drought HeatDifferent treatments

Prestress6th day

Leaf

tem

pera

ture

(∘C)

∗∗ ∗∗∗

∗∗∗p < 0.01∗∗p < 0.03∗p < 0.05

Drought + heat

(a)

20.16 20.68 20.32 20.1918.29

22.23 23.11 23.36

0.005.00

10.0015.0020.0025.00


Prestress6th day

Leaf

tem

pera

ture

(∘C)

Control Drought Heat

∗∗∗∗∗

∗∗∗p < 0.01∗∗p < 0.03∗p < 0.05

Drought + heat

(b)

Figure 4: Bar diagrams showing temperature profiles of leaves between Day 0 (prestress) and Day 6 in two soybean cultivars Surge (a)and Davison (b). 𝑥-axis defines various abiotic stress treatments (drought, heat, and drought + heat) including control and 𝑦-axis definesmeasurement of leaf temperature OC. Each value represents the mean ± SE of three replicates and the asterisks designate the significance ofchanges from their subsequent control (∗∗∗𝑝 < 0.01, ∗∗𝑝 < 0.03, and ∗𝑝 < 0.05).

expressed proteins in both soybean varieties exposed todifferent stress conditions.

3.2.1. Genotypic Comparison at Molecular Levels for DifferentStresses. For the genotypic comparisons between Surge andDavison soybean varieties, we examined three different stressconditions (heat, drought, and drought plus heat) and a con-trol condition. Under drought stress conditions, out of the 44differentially expressed proteins in Davison, 16 proteins wereupregulated and 28 proteins were downregulated compared

to Surge. Under heat stress conditions, 19 proteins wereupregulated and 25 proteins were downregulated in Davisoncompared to Surge. Furthermore, when the soybean leaveswere exposed to combined drought plus heat stress, 21 pro-teins were upregulated and 23 proteins were downregulatedin Davison compared to Surge. To determine how these soy-bean leaf proteins vary under specific stress conditions, Venndiagramanalysis of the number of differentially expressed leafproteins was conducted (Figures 6(a) and 6(b)).


pI

150kDa

110 kDa100 kDa

70kDa

50kDa

40kDa

30kDa

20kDa

4 5.0 6.0 7.0

pH

Mol

ecul

ar w

eigh

t (M

W)

Figure 5: A representative figure showing 2D-DIGE analysis of differentially expressed soybean leaf proteins in response to heat, drought,and drought plus heat stresses. For each sample set, 30 𝜇g of sample protein was mixed with 1.0𝜇L of diluted Cy3 or Cy5 dye for labeling. Apooled protein sample containing equal amounts of all samples was labeled with Cy2. Labeled samples were subjected to isoelectric focusing(pH 4–7) followed by 12% SDS-polyacrylamide gels. Gel images were scanned using Typhoon TRIO (GE Healthcare, PA, USA) and wereanalyzed using Image QuantTL software (version 6.0, GE Healthcare, USA) and then subjected to in-gel analysis and cross-gel analysis usingDeCyder software, version 6.5 (GE Healthcare). The spot numbering on the merged 2D-DIGE image indicates the protein spots that werefound statistically significant between various treatments for mass spectrometric identification.

Hierarchical clustering analysis was performed on theexpression data for the 92 differentially expressed proteinspots in Surge compared to Davison under control, heatstress, drought stress, and drought plus heat stress conditionsto identify trends (Figure 7). Clustering analysis revealedthe assignment of the 92 protein spots into 6 prominentclusters (I–VI). Cluster I contained two proteins that weredownregulated in leaves when exposed to drought, anddrought plus heat conditions, but were upregulated underheat stress conditions. Cluster II was comprised of leafproteins that were downregulated under drought conditionsbut were not altered under the other conditions. ClusterIII consisted of proteins that were upregulated in responseto heat stress to a lesser extent than those in cluster I.Cluster IV included proteins that were not significantlyaltered in response to stress. The expression of protein spotsin cluster V displayed remarkable variability; under heat

stress conditions, these proteins were downregulated, butunder combined drought and heat stress conditions, theseproteins were upregulated. The proteins included in clusterVI displayed prominent downregulation under heat stressconditions but upregulation under combined drought andheat stress conditions.

3.2.2. Comparison of the Effects of Stress on DifferentGenotypes. To examine the molecular changes that occurunder stress conditions, the leaf proteomes of both vari-eties were analyzed (Figure 6(c)). Under drought stressconditions, it was found that 11 proteins (stromal 70 kDaheat shock-related protein, ATP-dependent zinc metallopro-tease, RuBisCO large subunit-binding protein subunit beta,elongation factor Tu, glutamine synthetase, photosystemII stability/assembly factor HCF136, malate dehydrogenase,fructose-bisphosphate aldolase, ferredoxin-NADP reductase,


24 15

15

18 2328

25Heat

DroughtHeat +

drought

(a) Proteins downregulated in Davison compared to Surge in different stressconditions

10

15 11

11

19Heat

16Drought

21

Heat +drought

(b) Proteins upregulated in Davison compared to Surge in different stressconditions

40

30

20

10

0

−10

−20

−30

−40

−26−20

−13

−31

−24 −22

Drought in Drought in Heat in Surge Heat in Davison Drought + heat(Surge) (Davison)

Drought + heatSurge Davison

1824

31

1320 22

Num

ber o

f pro

tein

s dow

n- o

r upr

egul

ated

(c)

Figure 6: Venn diagram showing the proteins which are unique to respective genotype and stress. The number of downregulated (a) andupregulated (b) proteins and comparison of two genotypes in relation to number of proteins that get affected as a result of various stresses(c).

and chlorophyll a-b binding protein 6A) were upregulated inDavison but were downregulated in Surge.

Under heat stress conditions, 13 proteins were down-regulated in Surge, whereas 31 proteins were downregu-lated in Davison compared to the control condition. Theresults revealed significant downregulation of 20 proteinsin Davison. The most downregulated proteins include stro-mal 70 kDa heat shock-related protein, ATP-dependent zincmetalloprotease FTSH 8, elongation factor Tu, and ribulosebisphosphate carboxylase large chain, but all of these proteinswere upregulated in Surge.

When the plants were exposed to drought plus heat stress,it was found that 22 proteins were downregulated and 22proteins were upregulated in Davison, whereas 24 proteinswere downregulated and 20 proteins were upregulated inSurge compared to plants under the control conditions.Under the drought plus heat stress conditions, 9 proteinswere

downregulated in Surge (the most downregulated proteinsinclude stromal 70 kDa heat shock-related protein, ATP-dependent zinc metalloprotease FTSH 8, ribulose bisphos-phate carboxylase small chain, and carbonic anhydrase 1).However, these proteins were upregulated in Davison.

3.3. Functional Correlation among the Identified Proteinsunder Drought and Heat Stress. To attain a better under-standing of the early proteomic responses, a majority ofthe differentially expressed proteins identified were classifiedinto various functional categories. This analysis (Figure 8(a))indicated that the 92 significantly altered protein spots iden-tified were found to be involved in different metabolic path-ways and processes, including photosynthesis (65.56%), ATPsynthesis (7.78%), protein biosynthesis (4.44%), superoxidedismutase activity (3.33%), protein folding (2.22%), lipidmetabolism (2.22%), photorespiration/response to hypoxia


Cluster I

Cluster II

Cluster IV

Cluster V

Cluster III

Cluster VI

404118612155542122141395279778672637368048517585435446410744546654750204976667271173161953888789215659607367695755586328433970832333130382990252468112327346278816142838482

Con

trol

Dro

ught

Hea

t

−3 0 3

Log2 ratio

Dro

ught+

heat

Figure 7: Heat map: Euclidean distance similarity metric and complete linkage method. Heat maps of proteins associated with soybeanplants (pooled lines) exposed to different stress conditions. Hierarchical clustering was done based on the log

2-transformed expression ratios

of protein spots using Gene Cluster 3.0 software with the Euclidean distance similarity metric and complete linkage method. Clusters werevisualized in JAVA TREEVIEW software.


(2.22%), respiration (2.22%), carbonate dehydratase activity(2.22%), acid phosphatase activity (2.22%), nitrogen fixation(1.11%), one-carbon metabolism (1.11%), calcium ion binding(1.11%), serine protease inhibitor (1.11%), and response to coldstress (1.11%).

An analysis was performed to elucidate the subcellularlocalization of the differentially expressed proteins (Fig-ure 8(b)). A large portion of the differentially expressedproteins identified were predicted to be chloroplastic (86%),whereas most of the remaining proteins were predicted tobe localized to the cytoplasm (5.55%), mitochondria (2.22%),vacuoles (2.22%), peroxisomes (1.11%), or ribosomes (1.11%).

3.4. Identified Proteins Affecting Photosynthesis under Droughtand Heat Stress Conditions. In this study, 25 proteins thatare directly or indirectly involved in photosynthesis werediscovered as differentially expressed in soybeans in responseto various stress conditions. Protein upregulation and down-regulation in Davison compared to Surge are presented inFigure 9.The proteomic analysis revealed that, under droughtstress, photosynthesis-related proteins, including ribulosebisphosphate carboxylase large chain (Spot 22), ribulose bis-phosphate carboxylase/oxygenase activase (Spot 41), trans-ketolase (Spot 40), sedoheptulose-1,7-bisphosphatase (Spot37), fructose-bisphosphate aldolase 1 (Spot 48), phospho-glycerate kinase (Spot 43), oxygen-evolving enhancer protein2-1 (Spot 74), chlorophyll a-b binding protein 3 (Spot 75),ATP-dependent zinc metalloprotease (Spot 9), ATP synthasesubunit alpha (Spot 17), ATP synthase subunit beta (Spot 19),photosystem II stability/assembly factor HCF136 (Spot 45),and ferredoxin-NADP reductase (Spot 54), were downreg-ulated in both soybean varieties (Figure 9). This indicatesthat drought stress favors a reduction in the activities ofphotosynthetic carbon reduction cycle enzymes, specificallyribulose-1,5-bisphosphate (RuBP) regeneration and inhibi-tion of RuBisCO activity [53–55].These findings also supportthe current physiological analysis of the photosynthesis.

The photosynthesis rate in higher plants depends on theactivity of RuBisCO and the regeneration of RuBP. It has beenreported that drought stress results in a rapid decrease inthe abundance of RuBisCO small subunit (rbcS) in tomato[56]. Similarly, our analysis also revealed a decreased (2–4-fold) synthesis of RuBisCO large and small chain enzymeunder drought stress which may have direct effect on pho-tosynthesis. The literature suggests that this inhibition ofRuBisCO activity was due to the binding of inhibitors suchas 2-carboxyarabinitol 1 phosphate (CA1P) to RuBisCO [57].However, the interactions between RuBisCO and inhibitorsare known to be prevented by ATP hydrolysis and RuBisCOactivase, which regulates the active site conformation ofRuBisCO, removes the inhibitors and the bound inactiveRuBP, and allows RuBisCO to undergo rapid carboxylation[58]. Furthermore, we found that, in response to droughtstress, RuBisCO activase is also downregulated by 2-foldunder stress conditions compared to the control conditions.This result suggests that, under drought stress, RuBisCO acti-vase cannot prevent the interaction between RuBisCO andits inhibitors and cannot remove the bound RuBP to activateRuBisCO. Along with the downregulation of RuBisCO and

RuBisCO activase, we found that, under drought stress, otherkey Calvin cycle enzymes that play a crucial role in car-boxylation were downregulated, including transketolase (2-fold compared to the control condition), sedoheptulose-1,7-bisphosphatase (1–2.4-fold), fructose-bisphosphate aldolase1 (1.5-fold), and phosphoglycerate kinase (4–17-fold). Thisresult suggests that carboxylation is also reduced due to theinhibition of these enzymes, thereby decreasing the levelof photosynthesis and ultimately affecting the growth anddevelopment of these soybean plants.

We found that, due to heat stress, crucial proteins thatregulate electron transport activity were also downregulatedincluding ATP synthase subunit alpha (2.18-fold in Surgecompared to Davison), ATP synthase subunit beta (1.74-fold in Davison compared to Surge), photosystem II stabil-ity/assembly factor HCF136 (1.15-fold in in Davison com-pared to Surge), oxygen-evolving enhancer protein 2-1 (1.8-fold in Surge compared to Davison), and ferredoxin-NADPreductase (1.57-fold in Davison compared to Surge). Theseresults suggest that, in both soybean cultivars, heat stressnegatively affects electron transport activity, including PSIIdownregulation, thereby reducing the level of photosynthesis[59].

3.5. Upregulation of EF-Tu during Heat Stress in Surge and theRole of EF-Tu in Protecting the PhotosynthesisMachinery. 2D-DIGE analysis revealed that the chloroplast protein synthesiselongation factor, EF-Tu protein (Spot 31), was upregulatedby 4.6-fold in Surge compared to Davison under heat stress.Additionally, we found that, in Surge, under heat stress, fewer(13) proteins were downregulated than under drought stress,whereas 31 proteins were downregulated in Davison underheat stress. Chloroplast EF-Tu is a protein that is involvedin the elongation of polypeptides during the translationalprocess, and it belongs to a nuclear-encodedmultigene family[60, 61]. The heat stress-induced accumulation of EF-Tuin mature plants is thought to protect the photosynthesismachinery [62]. It has been reported that maize EF-Tu playsa crucial role in heat tolerance by acting as a molecularchaperone, and it has been found to protect heat-labile citratesynthase, RuBisCO activase, and malate dehydrogenase fromthermal accumulations [63, 64]. The results indicate that,upon heat stress, EF-Tu assembles in the cytosol in Surge(as seen by an increase in the protein levels of Spots 29–31). Therefore, based on these results, EF-Tu could serve asa biomarker of heat stress in soybeans.

Bioinformatics analysis of protein-protein interactionsof this study revealed that the EF-Tu protein undergoeschaperone-mediated interaction with all photosynthesis-related enzymes, including RuBisCO activase, malate dehy-drogenase, phosphoglycerate kinase, and sedoheptulose-1,7-bisphosphatase (Figure 12). An interaction between cpn60𝛽and RuBisCO activase has been reported in Arabidopsisfor acclimating photosynthesis to heat stress [65]. Conse-quently, it was found that, in Surge variety under heat stress,the expression of most proteins related to photosynthesiswas conserved, that is, either upregulated or unchanged;in particular, RuBisCO activase (Spot 41) was upregulated8.71-fold in Surge compared to Davison. We propose that


65.56%

7.78%

4.44%

3.33%

2.22%

2.22%

2.22%

2.22%

2.22%

2.22%

1.11%

1.11%

1.11%

1.11%

1.11%

0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00

Photosynthesis

ATP synthesis

Protein biosynthesis

Superoxidedismutase activity

Response to stress/protein folding

Lipid metabolism

Photorespiration/response to hypoxia

Respiration

Carbonatedehydratase activity

Acid phosphataseactivity

Nitrogen fixation

One-carbonmetabolism

Calcium ion binding

Serine proteaseinhibitor

Response to coldstress

Proteins (%)

(a) Functional groups of differentially expressed proteins identified in the seedling

86%

5.55%

1.11%

2.22%

1.11%

1.11%

2.22%

0 10 20 30 40 50 60 70 80 90 100

Chloroplastid

Cytoplasm

Peroxisome

Mitochondria

Extracellular region

Ribosome

Vacuole

Proteins (%)

(b) The predicted subcellular localization of the identified proteins

Figure 8:Graphical representation of functional categories of identified proteins.𝑥-axis defines various functional groups and𝑦-axis indicates% of proteins. B-Mapping for the subcellular localization of the identified proteins. 𝑥-axis defines various subcellular organelles and 𝑦-axisindicates % of proteins. Numerical values symbolize the percentage of proteins in each functional class.


−5

−4

−3

−2

−1

0

1

2

−4

−3

−2

−1

0

1

3

2

−4

−5

−3

−2

−1

0

1

3

2

−1.5

−2

−2.5

−1

−0.5

0.5

0

1

1.5

−1.5

−1

−0.5

0.5

0

1

1.5

2

−15

−20

−10

−5

5

0

−1.5

−2

−1

−0.5

0

−1.5

−1

−0.5

0.5

0

1

1.5

−1.5

−2

−1

−0.5

0.5

0

1

1 2 3 5 64

1.5

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Prot

ein

ratio

Ribulose bisphosphate carboxylase large chain (Spot 22)

Ribulose bisphosphate carboxylase/oxygenase activase (Spot 41)

Transketolase (Spot 40) Sedoheptulose-1,7-bisphosphatase (Spot 37)

Fructose-bisphosphate aldolase 1 (Spot 48)Phosphoglycerate kinase (Spot 43)

ATP synthase subunit alpha (Spot 17)

Photosystem II stability/assembly factor HCF136 (Spot 45)

Ferredoxin-NADP reductase (Spot 54)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

D/C

(Dv)

D/C

(Sur

)

H/C

(Dv)

H/C

(Sur

)

D+

H/C

(Dv)

D+

H/C

(Sur

)

H/C

(Sur

)

D+

H/C

(Dv)

Figure 9: Bar diagrams showing differential regulation of photosynthesis-related proteins. This response is to various stress treatments intwo soybean cultivars Surge and Davison. Dv = Davison; Sur = Surge; D = drought, H = heat, and D + H = drought plus heat stress. Theprotein abundance is presented as protein ratio (𝑦-axis) compared to control in response to various abiotic stress treatments (drought, heat,and drought + heat) on 𝑥-axis.


62.67

339.62

48.50 40.70

0.0050.00

100.00150.00200.00250.00300.00350.00400.00450.00

Different treatmentsControl Drought Heat Drought + heat

Hyd

roge

n pe

roxi

de co

ncen

trat

ion

(nm

ol/𝜇

L)

∗∗∗∗∗p < 0.01∗∗p < 0.03∗p < 0.05

(a)

63.77

404.70

114.9148.09

0.0050.00

100.00150.00200.00250.00300.00350.00400.00450.00500.00


Hyd

roge

n pe

roxi

de co

ncen

trat

ion

(nm

ol/𝜇

L)


∗∗∗∗

∗∗∗p < 0.01∗∗p < 0.03∗p < 0.05

(b)

Figure 10: Bar diagram showing hydrogen peroxide levels under various stress treatments among Surge (a) and Davison (b). 𝑥-axis definesvarious abiotic stress treatments (drought, heat, and drought + heat) including control and the 𝑦-axis defines measurement of hydrogenperoxide concentration in nmol/𝜇L. Each value represents the mean ± SE of five replicates and the asterisks designate the significance ofchanges from their subsequent control (∗∗∗𝑝 < 0.01, ∗∗𝑝 < 0.03, and ∗𝑝 < 0.05).

EF-Tu is synthesized upon heat stress and protects theheat-induced degradation of all photosynthesis-related pro-teins and enzymes, especially RuBisCO activase, maintainingthe photosynthesis levels.

3.6. Role of Stromal Heat Shock-Related Protein during SevereHeat Stress. Based on 2D-DIGE analysis, we found thatthe stromal 70 kDa heat shock-related protein (Spot 8) wasupregulated in Surge by 4.15-fold compared to Davison(Figure S6), and, compared to the control conditions, it wasupregulated by 2.33-fold in Surge under heat stress condi-tions. This molecular chaperone maintains cellular home-ostasis in cells under adverse abiotic stress conditions. It hasbeen found that stromal 70 kDa heat shock-related proteinof Arabidopsis thaliana is responsible for protein folding andcan assist in protein refolding under heat stress conditions[66].The results also indicate that, under heat stress, proteinsrelated to the photosynthesis machinery are protected. Takentogether, the analysis indicates that stromal 70 kDa heatshock-related protein plays a role in the maintenance orbiogenesis of chloroplast proteins, thereby maintaining thelevel of photosynthesis [20, 67].

A further bioinformatics analysis of protein-proteininteractions revealed that this HSP protein undergoes two-way interactions, one with a molecular chaperone (GrpE)and one directly with phosphoglycerate kinase. As discussedabove, under heat stress, most proteins related to photo-synthesis are more abundant in Surge than in Davison.Thus, we propose that stromal 70 kDa heat shock-relatedproteinmight activate amultiprotein interaction cascade thatmaintains the biogenesis of chloroplast proteins under heatstress and protects against the heat-induced degradation ofall photosynthesis-related proteins and enzymes, preservingthe photosynthesis levels.

3.7. Quantification of ROS and the Role of Carbonic Anhydrasein Plant Protection. Studies have shown that increasing levelsof water stress in Vigna plants may increase ROS levels

by means of hydrogen peroxide [28]. Results of presentstudy (Figure 10) indicate a 5.4-fold elevation in the H

2O2

level under drought stress compared to the control condi-tions in Surge (62.67 nmol𝜇L−1 to 339.62 nmol𝜇L−1) anda nearly 6.3-fold elevation in Davison (63.77 nmol𝜇L−1 to404.70 nmol𝜇L−1) on the sixth day of stress. Under heatstress, we found no significant difference in the H

2O2level

in Surge, but in Davison, the H2O2level was increased by 1.8-

fold.The proteomic analysis shows that carbonic anhydrase

1 (Spot 72) is upregulated by 1.8-fold due to drought stressand by 1.6-fold due to combined drought plus heat stress inDavison (Figure S7). Carbonic anhydrase is a zinc-containingmetalloenzyme, and the specific association between car-bonic anhydrase and RuBisCO enables CO

2to interact

with RuBisCO and maintains the functional machinery ofRuBisCO [68]. Under drought stress conditions, when plantsdetect a limitation of water availability, stomatal closure istriggered, which limits the entrance of CO

2, resulting in a

net reduction of photosynthesis. This confers increased ROSaccumulation with oxidative stress [69]. It has been reportedthat higher expression of carbonic anhydrase in the cellincreases its resistance to cytotoxic concentrations of H

2O2

[70].Thus, it is possible that, under drought stress conditions,the upregulated expression of carbonic anhydrase in Davisonprincipallymakes the plant cells resistant to toxic H

2O2levels

and protects the plant from oxidative stress. Although wedetected a higher level of hydrogen peroxide in drought-stressed Davison plants, no significant phenotypic changeswere detected on the sixth day of stress.

3.8. Quantification of ABA under Different Abiotic StressConditions. ABA is a phytohormone critical for plant growthand plays a key role in integrating various stress signals [71].Several studies suggested that, in response to the water andhigh temperature stress, stomatal movement is controlled byABA signaling [72]. In this study, we quantified the ABAconcentration in the leaf under all the stress conditions


169.92 417.83

1,233.56

4,467.84

50.91528.39

2,995.31

6,766.94

0.00

1,000.00

2,000.00

3,000.00

4,000.00

5,000.00

6,000.00

7,000.00

8,000.00

9,000.00

10,000.00

ABA

conc

entr

atio

n (p

mol

/mL)

< 0.03

< 0.05

Surg

e con

trol

Surg

e hea

t

Surg

e dro

ught

Surg

e (dr

ough

t+he

at)

Dav

ison

cont

rol

Dav

ison

heat

Dav

ison

drou

ght

Dav

ison

(dro

ught

+he

at)

∗

∗

∗∗

∗∗

∗∗∗

∗∗∗

∗∗∗p∗∗p∗p

< 0.01

Figure 11: Bar diagram showing quantification of ABA in different abiotic stress conditions in Surge and Davison. 𝑥-axis defines variousabiotic stress treatments (drought, heat, and drought + heat) including control and the 𝑦-axis defines measurement of ABA concentration(pmol/mL). Each value represents themean± SE of five replicates and the asterisks designate the significance of changes from their subsequentcontrol (∗∗∗𝑝 < 0.01, ∗∗𝑝 < 0.03, and ∗𝑝 < 0.05).

(Figure 11). We found that, compared to the control plants,the highest level of accumulation of ABA occurred duringthe drought plus heat stress conditions. In Surge, it was 26-fold higher, and in Davison, it was 132-fold higher comparedto the control conditions. In the drought-stressed leaves, wefound ABA levels to be 7-fold higher in Surge and 58-foldhigher in Davison compared to the control plants. Underheat stress, ABA is upregulated by 2-fold in Surge and 10-fold in Davison compared to the normal plants. Throughinvestigating the change of abscisic acid (ABA) levels, wefound that drought plus heat stress has the highest level ofABA accumulation, which also correlates with the stomatalconductance measurements.

3.9. The Identified Protein “Glutamine Synthetase” and ItsAssociation with Nitrogen Metabolism. Based on the analy-sis of the soybean leaf proteome, we found that cytosolicglutamine synthetase leaf isozyme (Spot 33) was downregu-lated by 4-fold in Davison under drought stress conditionscompared to Surge. This enzyme catalyzes a major reactionof nitrogen metabolism, that is, the assimilation of ammo-nium to glutamine using the substrate glutamic acid [73].Reduction of glutamine synthetase expression under droughtconditions has been reported as a protective mechanismfor plants because the intermediate nitric oxide is an activeradical.Thus, we predict that the same protective mechanismalso occurs in soybean leaves [21, 74]. Additionally, it hasbeen reported that the overexpression of cytosolic glutaminesynthetase in poplar enhances the photorespiration underdrought stress and that it also contributes to the protectionof photosynthesis [75, 76].

3.10. Protein-Protein Interaction Analysis. For any systems-level understanding of cellular functions, it is crucial toappropriately elucidate all functional interactions betweenthe proteins in the cell at a given time. We searched fora variety of functional protein-protein interactions betweenthe differentially expressed proteins in soybeans in responseto abiotic stress to obtain an improved understanding ofthese protein-protein interactions, including stable com-plexes, metabolic pathways, and a wide range of regula-tory interactions [44] (Figure 12 and Table 1). From thenetwork, we discovered 10 major proteins (most signifi-cantly, ribulose bisphosphate carboxylase, ribulose bispho-sphate carboxylase/oxygenase activase, ATP synthase sub-unit alpha/beta, sedoheptulose-1,7-bisphosphatase, photosys-tem II stability/assembly factor HCF136, ferredoxin-NADPreductase, elongation factor Tu, and carbonic anhydrase 1)having more than 9 interactions that are considered to beat the central body of the network. Drought or heat stressmediated downregulation or upregulation of those centralbody proteins of the network may also affect their predictedpartners, which ultimately could affect the molecular signal-ing by collapsing the whole cellular network.

4. Discussion

Drought and high temperature stress conditions cause exten-sive losses to crops’, including soybean, productionworldwide[77, 78]. Individually, the effect of drought or heat stressconditions on soybeans has been the subject of intenseresearch [79, 80], but to the best of our knowledge, no suchdetailed proteomics-based study attempt has been made to


Table 1: Data used to predict protein-protein network. Protein spot numbers are corresponding to the 2D-DIGE gel and respective massspectrometry based protein identifications are given as UniProt ID.

Spot number Protein name UniProt accession number1 Acyl-[acyl-carrier-protein] desaturase 6 Q0J7E42 Lipoxygenase Q434403 Ribulose bisphosphate carboxylase large chain Q018734 Transketolase, chloroplastic O202507 Stromal 70 kDa heat shock-related protein Q020289 ATP-dependent zinc metalloprotease FTSH 8 Q8W58513 RuBisCO large subunit-binding protein subunit beta P0892716 ATP synthase subunit alpha Q2PMS819 ATP synthase subunit beta Q2PMV028 Ribulose bisphosphate carboxylase/oxygenase activase Q4028129 Elongation factor Tu, chloroplastic Q4346733 Glutamine synthetase leaf isozyme P1510234 S-adenosylmethionine synthase 4 A7PRJ635 Probable calcium binding protein CML33 Q9SRP437 Sedoheptulose-1,7-bisphosphatase O2025242 Phosphoglycerate kinase P1278244 Photosystem II stability/assembly factor HCF136 O8266046 Fructose-bisphosphate aldolase 1 Q0151649 Malate dehydrogenase 1 Q9ZP0650 Trypsin inhibitor 1 Q4366752 Chloroplast stem-loop binding protein of 41 kDa b Q9SA5253 Ferredoxin-NADP reductase, leaf isozyme 1 Q9FKW6-156 Oxygen-evolving enhancer protein 1 P1422658 Chlorophyll a-b binding protein of LHCII type I P0822159 Chlorophyll a-b binding protein Q10HD060 Chlorophyll a-b binding protein 13 P2748961 2-Cys peroxiredoxin BAS1-like Q9C5R862 Ribulose bisphosphate carboxylase small chain PW9 P2666763 Ribulose bisphosphate carboxylase small chain 2A P2657564 Superoxide dismutase [Fe] P2875966 Carbonic anhydrase 1 P4651267 RuBisCO-associated protein P3965771 Stem 31 kDa glycoprotein P1074373 Stem 28 kDa glycoprotein P1549074 Oxygen-evolving enhancer protein 2-1 Q7DM3976 Chlorophyll a-b binding protein 6A P1236077 Ribulose bisphosphate carboxylase large chain (fragment) P2841680 Ribosomal protein L7/L12 Q4BZ0681 Glycine cleavage system H protein 1 P2585587 Ribulose bisphosphate carboxylase small chain 1 P00865

date.The cooccurrence of drought plus heat stress is of specialinterest because they occur together in the field. Recent stud-ies have revealed that the responses of plants to two differentabiotic stresses that occur simultaneously are exclusive andcannot be directly inferred from the response of plants toeach of the different stresses applied independently [81].Thus,to emphasize the molecular, physiological, and proteomicaspects of stress combination and to unravel the underlyingmechanisms of soybean responses to drought, heat stress,and cooccurring stresses, we performed a proteomic study

combined with physiological and computational analysis tofacilitate the discovery of genes that can enhance the tolerancecapabilities of soybean crops to the stress conditions [82].

Reduced stomatal conductance is one of the most sensi-tive responses to drought stress, observed in kidney beans andvarious C3 plants [48, 83]. At the level of soybean physiology,we concluded that drought stress, heat stress, and droughtplus heat stress reduce stomatal conductance, which is causedby decreasing the leaf water potential via transpiration ratealterations in both varieties. Stomatal conductance is also


4

53

80

44

59

58

60

71

73

33

64

2

35 1

62

63

63

6287

80

5077

19

29

16

4997

74

13

56

34

66

6633

52

81

61

46

76

37

67

87

28

3

42

Color interaction (Photosynthesis

I)

Lipid metabolismATP synthesisProtein biosynthesis

Acid phosphataseCarbonate dehydrataseCold stress responseRespiration

No interaction1 ≤ I ≤ 4

5 ≤ I ≤ 9

I > 9

Figure 12: Protein-protein interaction network predictions made by bioinformatics analysis. The prediction is between differentiallyexpressed proteins due to different stress conditions and identified by proteomic study. The network comprises 46 nodes and 139 edges.Symbol color corresponds to degree of interactions (color code on the left side). Symbol shapes indicate specific protein function. Nodelegends indicate the spot number corresponding to the particular protein spot from 2D-DIGE gel. Follow Table 1 for protein names forcorresponding spot number [note: duplicate insertion of few spot numbers (Spot numbers 33, 62, 63, 66, and 87) is incorporated to avoidovercrowding of the network. Spot number 80 appears two times with two different shapes indicating that the same protein is counted fortwo different functions].

reliant on leaf temperature via transpiration rate, and weobserved that the increase of leaf temperature is correlatedto the stomatal conductance and transpiration rates. Pho-tosynthesis is among the primary processes affected by thedrought and heat stresses [16, 84, 85]. As soil moisturedecreases during drought stress and cooccurring droughtand heat stress, the water content of mesophyll tissue alsoreduces, ultimately affecting the photosynthetic physiology,principally carbon assimilation and energy use of the plant.Surprisingly, there was no such significant effect observedduring heat stress alone, which actually establishes the factthat consequences of water deficiency have an enormousimpact on altering the photosynthetic machinery [86].

Gel-based proteomic separation is widely used in variousplants’ studies to decipher abiotic stress-responsive mech-anisms for its high precision exclusively for comparativeproteomics [87]. Likewise, we used a gel-based proteomicapproach to elucidate the mechanisms underlying the earlyresponses of soybeans to various abiotic stresses. To propose afunctional relationship among proteins in response to abioticstress, we performed functional categorization and subcel-lular localization of various differentially expressed proteins[88, 89]. Our proteomic analysis revealed that a total of 44proteins were significantly changed in soybean leaves after 6days of different stress exposures (drought, heat, and droughtplus heat), and their functional categorization showed that

most of the differentially changed proteins were related tophotosynthesis, ATP synthesis, and protein biosynthesis.Thesubcellular localization study reveals that most of the pro-teins are actively synthesized in chloroplast and cytoplasm.We found that 25 proteins related to photosynthesis weredownregulated during stress conditions in both the soybeanvarieties. Most importantly, downregulation of RuBisCO andRuBisCO activase enzymes during drought stress reduces thecarboxylation process because the limitation of the RuBisCOactivase prevents the reactivation of RuBisCO molecules[53, 90]. Previous reports on Phaseolus vulgaris and droughtstress showed that the Calvin cycle enzymes’ activities areaffected significantly [91]. Similarly, our study also indicatesthat, under drought stress, key enzymes ofCalvin cycle—suchas transketolase, sedoheptulose-1,7-bisphosphatase, fructose-bisphosphate aldolase 1, and phosphoglycerate kinase—thatplay major roles in carboxylation are also downregulated,resulting in reduced levels of photosynthesis, thus affectingthe development and growth of the soybean plant. Figures 13and 14 show models for how the photosynthesis level couldbe reduced after drought/heat stress via alternated proteinexpressions and protein-protein interactions. It was foundthat salt stress-induced effects in cowpea plants negativelyaffect nitrogen assimilation and overexpressed glutaminesynthetase gene in rice modifies nitrogen metabolism duringabiotic stress [92, 93]. Surprisingly, our soybean leaf proteome


CA1P RuBisCOlarge chain

Small chain 1

Small chain PW9

Small chain

2A

Normal condition

RuBisCOactivase

ATP-dependent zinc metalloprotease

Normal level of photosynthesis and growth

Activation of RuBisCO by removing bound RuBP RuBP

Block

DownregulationUpregulation

(a)

RuBisCOactivase

ATP-dependent zinc metalloprotease

RuBisCO large chain

Reduced RuBP

regeneration

Drought condition

Drought condition

Drought condition

SBPase

Aldolase

Transketolase

Calvin cycle and

PPP

Reduced level of photosynthesis and retarded growth

RNP7

No removal of bound RuBP, thus RuBisCO inactivated

Phosphoglyceratekinase

DownregulationUpregulation

BlockEnzyme having less effect

Small chain 1

Small chain PW9

Small chain

2A

CA1PRuBP

26S proteasome

(b)

Figure 13: Predicted analysis of the mechanism by which the photosynthesis rate is reduced. (a) In normal condition, RuBisCO activaseremoves the RuBP from RuBisCO and photosynthesis is not affected. (b) During drought stress, RuBisCO activase and ATP-dependentzinc metalloprotease are downregulated and CA1P (2-carboxyarabinitol 1 phosphate, a potent inhibitor of RuBisCO) is upregulated; thus noremoval of RuBP from RuBisCO occurs, and with that the Calvin cycle and pentose phosphate pathway (PPP) enzymes are downregulatedwhich ultimately results in reduced level photosynthesis and retarded growth.

analysis revealed that cytosolic glutamine synthetase leafisozyme, which is responsible for assimilation of ammoniumto glutamine using substrate glutamic acid, was highly down-regulated in Davison as compared to Surge during droughtstress conditions, resulting in negatively affected nitrogenassimilation.

Studies have shown that heat stress reduces yields by lim-iting electron transport activity in cotton [94] andArabidopsis[95]. The physiological and proteomic data of our studyreveal that the PSII could be damaged during heat stress, andthis damage reduces net photosynthesis levels in soybeans.

Crucial proteins—for example, ATP synthase, HCF136,oxygen-evolving enhancer protein 2-1, and ferredoxin-NADPreductase that regulate the electron transport activity—werefound to be downregulated during heat stress. Figure 14shows the predicted analysis of irreversible inhibition ofphotosynthesis during heat stress and the protein-proteininterplay that negatively affects the electron flow. Duringheat stress, we observed that the EF-Tu protein is highlyupregulated in Surge, which is consistent with the earlierfindings in heat tolerant maize [96]. So we predict that highlevels of heat stress-induced expression of EF-Tu in Surge


eFNR

NADPHATP

e

Normal condition

Photosystem IIe

α

Photosystem I

e

Fd ADP + Pi𝛽 𝛼NADP+ + H+

Oxygen-evolvingenhancer

CAB1

HCF136

H+ + O2

H2O

(a)

αHCF136

enhancer

eFNR

Heat stress

ATP

Downregulation

Block

Enzyme having less effect

Carbonfixation

Heat stress

NADPH

CAB1

Photosystem II Photosystem I

e

FdADP + Pi

𝛽 𝛼

NADP+ + H+

H+ + O2

H2O

Oxygen-evolvingenhancer

e

(b)

Figure 14: Predicted heat stress and electron transport block mechanism. Predicted analysis of irreversible inhibition of photosynthesisduring heat stress and the protein-protein interplay that negatively affects the electron flow. (a) In normal condition, there is normal electronflow and no reduction in carbon fixation. (b) Due to heat stress effect, all key proteins that mediate electron flow are downregulated and blockthe electron flow, and as a result downregulation of ATP synthase reduces the production of ATP and as a result carbon fixation process slowsdown.

activate the heat stress tolerance mechanisms, which mightactually be a safeguard for the heat-labile proteins such ascitrate synthase, RuBisCO activase, and malate dehydroge-nase from thermal degradation; and finally, this action atthe cellular level enables the protection of photosyntheticmachinery. In support of the heat tolerance mechanismin Surge, we found a higher accumulation of a stromal70 kDa HSP, which we predict might regulate functionallyvital processes crucial for the plant’s survival. This findingis relevant to the earlier reports reporting that HSP familyof chaperons can promote the protein refolding to maintainprotein homeostasis under heat stress conditions [97]. Thephysiological study also indicates that, in Surge, there is

no significant change of photosynthesis levels due to heatstress.

One of the inevitable outcomes from the biochemicalassays is the increase of ROS level (hydrogen peroxide)during drought stress in both Surge and Davison. We alsofound higher amounts of carbonic anhydrase accumulationin the cell which probably aid the cell in becoming moreresistant to cytotoxic concentrations of H

2O2in drought-

stressed Davison plants; thus, significant changes at thephenotypic level were not noticed at the sixth day of thestress [70]. While investigating the changes in ABA levelsduring stress, we found that the cooccurrence of droughtand heat stresses causes the highest accumulation of ABA in


soybeans compared to control plants, which is consistent withearlier findings in brassica and rice [98, 99].This observationalso correlates with the stomatal conductancemeasurements,statistical analysis by scatter plot of ABA concentration,and stomatal conductance measurement that shows a linearrelationship with 𝑅2 > 0.95, which indicates that an increasein ABA levels and a decrease in stomatal conductance havea kind of cause and effect relationship. While comparingthe ABA measurement analysis and stomatal conductancemeasurement, we found that, in cooccurring drought andheat stress, stomatal conductance value is lowest while ABAlevels are at the highest compared to drought-stressed andheat-stressed plants. After correlating the two observations,we concluded that the more severe the stress is, the more theABA synthesis ultimately results in the stomatal closure, thatis, lesser stomatal conductance. In response to the drought,heat, and cooccurring drought plus heat stresses, calciumlevels elevate as probable calcium binding protein CML33accumulates at higher levels [100]. These consequences acti-vate the phosphoprotein cascade, which ultimately activatesthe transcription of ABA biosynthesis precursors and finallysynthesizes ABA that aids in stomatal closure to reduce thetranspiration levels, CO

2assimilation, and photosynthesis for

the sake of survival of the plant under adverse environmentalconditions [71].

Finally, for the better understanding of the crosstalkbetween complex sets of abiotic stresses responsive pro-teins, we applied system biology approaches to determinethe holistic outlook of the molecular responses [17, 101].Our computational analysis of protein-protein interactionnetworks identifies 10 major proteinsthat havemore than 9interactions considered to be at the central body of thenetwork system. This protein network forecasts that droughtor heat stress mediated downregulation or upregulation ofthose central body proteins of the network may also affecttheir predicted partners, which ultimately could affect themolecular signaling by collapsing the whole cellular network.This systems biology study will assist in the breeding ofmore abiotic stress tolerant soybean varieties having highnutritional value.

Proteomics study on rice by Kim et al. has shown thatproteomic data can be used for crop improvement whicheventually leads to food security [42]. A clearer understand-ing of differential expression of various stress-related proteinsand inclusion of this knowledge to breed better varietiesmay trigger a speedy improvement of crop plants [26, 42].The extensive application of such quantitative proteomicapproaches together with mapping of posttranslational mod-ifications will provide us with comprehensive insights of theregulation of various stress-responsive proteins in complexbiological systems [102]. The agronomical perspective of thecurrent study will support the soybean breeders to developheat and drought tolerant soybean varieties.

Taken together, our study results identify the differentialexpression of a number of drought-, heat-, and cooccurringdrought and heat-responsive proteins. The observed changesin leaf protein expressions suggest that the regulation ofvarious molecular processes and signaling alters due todrought and heat stresses.

5. Conclusions

This study shows how drought, heat, and cooccurringdrought and heat stresses alter the soybean leaf proteome.Differential expression of 44 abiotic stress-responsive pro-teins suggests that various signaling cascades and molecularprocesses are affected due to drought and heat stresses.Furthermore, the result suggests that many differentiallyexpressed photosynthesis-related proteins perturb RuBisCOregulation, electron transport, and Calvin cycle during abi-otic stress. These findings, as well as upregulation of EF-Tu and higher level expressions of HSP, will help in betterunderstanding of the heat tolerance mechanisms in thesoybean varieties. The biochemical and proteomic assaysalso explain how higher amount of carbonic anhydraseaccumulation in the cell alleviates cytotoxic concentrationsof H2O2during drought stress. Moreover, we found that

cooccurring drought and heat stresses have the highest levelof ABA accumulations in Davison, which also correlates withour stomatal conductancemeasurements. Taken together, theresults presented provide a proteomic level explanation forthe abiotic stress effects on physiological processes of soybeanplants during abiotic stress conditions.

Abbreviations

% VWC: Percentage volume water content2D-DIGE: Two-dimensional differential in-gel

electrophoresisABA: Abscisic acidCA1P: 2-Carboxyarabinitol 1 phosphateCAB1: Chlorophyll a-b binding protein 1CML33: Probable calcium binding proteinCy2/Cy3/Cy5: Cyanine dye 2, 3, or 5DMF: N, N-DimethylformamideEF-Tu: Elongation factor TuFd: FerredoxinFNR: Ferredoxin-NADP+ oxidoreductaseHRP: Horseradish peroxidaseHSP: Heat shock-related proteinsPPP: Pentose phosphate pathwayRH: Relative humidityRNP7: Protein RNP-7ROS: Reactive oxygen speciesRuBisCO: Ribulose bisphosphate carboxylase

oxygenaseSTRING: Search tool for the retrieval of

interacting genes.

Conflict of Interests

The authors declare that they have no competing interests.

Authors’ Contribution

The idea was conceived by Jai S. Rohila. Aayudh Das,Moustafa Eldakak, Bimal Paudel, Homa Hemmati, andChhandak Basu did the experiments. Aayudh Das, MoustafaEldakak, Bimal Paudel, Dea-Wook Kim, Homa Hemmati,


Chhandak Basu, and Jai S. Rohila did the data analyses.Aayudh Das, Chhandak Basu, and Jai S. Rohila wrote thepaper.

Acknowledgments

The authors thank Dr. Senthil Subramanian for his assistancein seed procurement, Dr. YajunWu for his critical discussionsand help with ROS measurements, and Mr. Nathan Serfling(Department of English, SDSU) for his help to improve theEnglish of this paper. DWK acknowledges all support fromCooperative Research Program for Agriculture Science &Technology Development (Project no. PJ00953802) of RuralDevelopment Administration, Republic of Korea. This workwas supported by South Dakota Soybean Research & Pro-motion Council, South Dakota State University AgriculturalExperiment Station, CRIS Project SD00H541-15, and theDepartment of Biology & Microbiology.

References

[1] P. H. Graham and C. P. Vance, “Legumes: importance andconstraints to greater use,” Plant Physiology, vol. 131, no. 3, pp.872–877, 2003.

[2] South Dakota Soybean Research and Promotion Council, “Soy-beans Make Big Economic Impact,” 2014, http://sdsoybean.org/All About Soy/Economic Impact.

[3] U.S.D.A, “Soybeans and Oil Crops,” August 2014, http://www.ers.usda.gov/topics/crops/soybeans-oil-crops/related-data-sta-tistics.aspx#.U9wFxfldU08.

[4] M. Eldakak, S. I. M. Milad, A. I. Nawar, and J. S. Rohila,“Proteomics: a biotechnology tool for crop improvement,”Frontiers in Plant Science, vol. 4, no. 35, pp. 1–12, 2013.

[5] M. Farooq, A. Wahid, N. Kobayashi, D. Fujita, and S. M. A.Basra, “Plant drought stress: effects, mechanisms and manage-ment,” Agronomy for Sustainable Development, vol. 29, no. 1, pp.185–212, 2009.

[6] K. Kosova, P. Vıtamvas, I. T. Prasil, and J. Renaut, “lant pro-teome changes under abiotic stress-contribution of proteomicsstudies to understanding plant stress response,” Journal ofProteomics, vol. 74, no. 8, pp. 1301–1322, 2011.

[7] D. Jespersen and B. Huang, “Proteins associated with heat-induced leaf senescence in creeping bentgrass as affected byfoliar application of nitrogen, cytokinins, and an ethyleneinhibitor,” Proteomics, vol. 15, no. 4, pp. 798–812, 2015.

[8] G.-T. Liu, L.Ma,W. Duan et al., “Differential proteomic analysisof grapevine leaves by iTRAQ reveals responses to heat stressand subsequent recovery,”BMCPlant Biology, vol. 14, article 110,2014.

[9] L. Rizhsky, H. Liang, and R. Mittler, “The combined effect ofdrought stress and heat shock on gene expression in tobacco,”Plant Physiology, vol. 130, no. 3, pp. 1143–1151, 2002.

[10] J. A. Rollins, E. Habte, S. E. Templer, T. Colby, J. Schmidt,and M. Von Korff, “Leaf proteome alterations in the context ofphysiological andmorphological responses to drought and heatstress in barley (Hordeum vulgare L.),” Journal of ExperimentalBotany, vol. 64, no. 11, pp. 3201–3212, 2013.

[11] P. P. Mohammadi, A. Moieni, S. Hiraga, and S. Komatsu,“Organ-specific proteomic analysis of drought-stressed soy-bean seedlings,” Journal of Proteomics, vol. 75, no. 6, pp. 1906–1923, 2012.

[12] V. M. Rodrıguez, P. Soengas, V. Alonso-Villaverde, T. Sotelo, M.E. Cartea, and P. Velasco, “Effect of temperature stress on theearly vegetative development ofBrassica oleracea L.,”BMCPlantBiology, vol. 15, no. 145, pp. 1–9, 2015.

[13] G.M.Ali and S. Komatsu, “Proteomic analysis of rice leaf sheathduring drought stress,” Journal of Proteome Research, vol. 5, no.2, pp. 396–403, 2006.

[14] Z. Peng, M. Wang, F. Li, H. Lv, C. Li, and G. Xia, “A proteomicstudy of the response to salinity and drought stress in anintrogression strain of bread wheat,” Molecular and CellularProteomics, vol. 8, no. 12, pp. 2676–2686, 2009.

[15] G. H. Salekdeh, J. Siopongco, L. J. Wade, B. Ghareyazie, and J.Bennett, “Proteomic analysis of rice leaves during drought stressand recovery,” Proteomics, vol. 2, no. 9, pp. 1131–1145, 2002.

[16] M.M. Chaves, J. Flexas, and C. Pinheiro, “Photosynthesis underdrought and salt stress: regulation mechanisms from wholeplant to cell,”Annals of Botany, vol. 103, no. 4, pp. 551–560, 2009.

[17] G. R. Cramer, K.Urano, S.Delrot,M. Pezzotti, andK. Shinozaki,“Effects of abiotic stress on plants: a systems biology perspec-tive,” BMC Plant Biology, vol. 11, article 163, 14 pages, 2011.

[18] J.-K. Zhu, “Salt and drought stress signal transduction in plants,”Annual Review of Plant Biology, vol. 53, pp. 247–273, 2002.

[19] G. R. Cramer, S. C. Van Sluyter, D. W. Hopper, D. Pascovici,T. Keighley, and P. A. Haynes, “Proteomic analysis indicatesmassive changes inmetabolismprior to the inhibition of growthand photosynthesis of grapevine (Vitis viniferaL.) in response towater deficit,” BMC Plant Biology, vol. 13, no. 49, pp. 1–22, 2013.

[20] W. Wang, B. Vinocur, O. Shoseyov, and A. Altman, “Role ofplant heat-shock proteins and molecular chaperones in theabiotic stress response,” Trends in Plant Science, vol. 9, no. 5, pp.244–252, 2004.

[21] C. Plomion, C. Lalanne, S. Claverol et al., “Mapping theproteome of poplar and application to the discovery of drought-stress responsive proteins,” Proteomics, vol. 6, no. 24, pp. 6509–6527, 2006.

[22] W. Li, Z. Wei, Z. Qiao, Z. Wu, L. Cheng, and Y. Wang,“Proteomics analysis of alfalfa response to heat stress,” PLoSONE, vol. 8, no. 12, Article ID e82725, 2013.

[23] S. Muneer, Y. G. Park, A. Manivannan, P. Soundararajan, and B.R. Jeong, “Physiological and proteomic analysis in chloroplastsof Solanum lycopersicum L. under silicon efficiency and salinitystress,” International Journal of Molecular Sciences, vol. 15, no.12, pp. 21803–21824, 2014.

[24] G. Cornic andA.Massacci, “Leaf photosynthesis under droughtstress,” in Photosynthesis and the Environment, N. R. Baker, Ed.,pp. 347–366, Springer, Amsterdam, The Netherlands, 1996.

[25] A. M. Timperio, M. G. Egidi, and L. Zolla, “Proteomics appliedon plant abiotic stresses: role of heat shock proteins (HSP),”Journal of Proteomics, vol. 71, no. 4, pp. 391–411, 2008.

[26] G. K. Agrawal, R. Pedreschi, B. J. Barkla et al., “Translationalplant proteomics: a perspective,” Journal of Proteomics, vol. 75,no. 15, pp. 4588–4601, 2012.

[27] C. A. Jaleel, K. Riadh, R. Gopi et al., “Antioxidant defenseresponses: physiological plasticity in higher plants under abioticconstraints,”Acta Physiologiae Plantarum, vol. 31, no. 3, pp. 427–436, 2009.

[28] S. P. Mukherjee and M. A. Choudhuri, “Implications of waterstress-induced changes in the levels of endogenous ascorbicacid and hydrogen peroxide in Vigna seedlings,” PhysiologiaPlantarum, vol. 58, no. 2, pp. 166–170, 1983.


[29] J. Mundy and N.-H. Chua, “Abscisic acid and water-stressinduce the expression of a novel rice gene,”The EMBO Journal,vol. 7, no. 8, pp. 2279–2286, 1988.

[30] I. E. Henson, “The heritability of abscisic acid accumulation inwater-stressed leaves of pearl millet [Pennisetum americanum(L.) leeke],” Annals of Botany, vol. 53, no. 1, pp. 1–12, 1984.

[31] X. Wang, T. Kuang, and Y. He, “Conservation between higherplants and the moss Physcomitrella patens in response to thephytohormone abscisic acid: a proteomics analysis,” BMC PlantBiology, vol. 10, article 192, 11 pages, 2010.

[32] M. Gong, Y.-J. Li, and S.-Z. Chen, “Abscisic acid-inducedthermotolerance in maize seedlings is mediated by calciumand associated with antioxidant systems,” Journal of PlantPhysiology, vol. 153, no. 3-4, pp. 488–496, 1998.

[33] P. R. Houser, W. J. Shuttleworth, J. S. Famiglietti, H. V. Gupta,K. H. Syed, and D. C. Goodrich, “Integration of soil moistureremote sensing and hydrologic modeling using data assimila-tion,” Water Resources Research, vol. 34, no. 12, pp. 3405–3420,1998.

[34] X. Ma, N. L. Sukiran, H. Ma, and Z. Su, “Moderate droughtcauses dramatic floral transcriptomic reprogramming to ensuresuccessful reproductive development in Arabidopsis,” BMCPlant Biology, vol. 14, article 164, 2014.

[35] N. B. A. Thu, X. L. T. Hoang, H. Doan et al., “Differentialexpression analysis of a subset of GmNAC genes in shootsof two contrasting drought-responsive soybean cultivars DT51andMTD720 under normal and drought conditions,”MolecularBiology Reports, vol. 41, no. 9, pp. 5563–5569, 2014.

[36] W. J. Hurkman and C. K. Tanaka, “Solubilization of plantmembrane proteins for analysis by two-dimensional gel elec-trophoresis,” Plant Physiology, vol. 81, no. 3, pp. 802–806, 1986.

[37] M. L. Robbins, A. Roy, P.-H. Wang et al., “Comparative pro-teomics analysis by DIGE and iTRAQ provides insight into theregulation of phenylpropanoids inmaize,” Journal of Proteomics,vol. 93, pp. 254–275, 2013.

[38] D. Gupta, M. Eldakak, J. S. Rohila, and C. Basu, “Biochemicalanalysis of ‘kerosene tree’ Hymenaea courbaril L. Under heatstress,” Plant Signaling and Behavior, vol. 9, no. 10, Article IDe972851, pp. 1–10, 2014.

[39] G. Poschmann, B. Sitek, B. Sipos et al., “Identification ofproteomic differences between squamous cell carcinoma ofthe lung and bronchial epithelium,” Molecular & CellularProteomics, vol. 8, no. 5, pp. 1105–1116, 2009.

[40] K. Aihara, K. J. Byer, and S. R. Khan, “Calcium phosphate–induced renal epithelial injury and stone formation: involve-ment of reactive oxygen species,” Kidney International, vol. 64,no. 4, pp. 1283–1291, 2003.

[41] B. G. de los Reyes, S. J. Yun, V. Herath et al., “Phenotypic,physiological, and molecular evaluation of rice chilling stressresponse at the vegetative stage,” in Rice Protocols, vol. 956 ofMethods in Molecular Biology, pp. 227–241, Humana Press, NewYork, NY, USA, 2013.

[42] S. T. Kim, S. G. Kim, G. K. Agrawal, S. Kikuchi, and R. Rakwal,“Rice proteomics: a model system for crop improvement andfood security,” Proteomics, vol. 14, no. 4-5, pp. 593–610, 2014.

[43] F. Sievers andD.G.Higgins, “Clustal omega, accurate alignmentof very large numbers of sequences,” in Multiple SequenceAlignment Methods, pp. 105–116, Springer, 2014.

[44] D. Szklarczyk, A. Franceschini, M. Kuhn et al., “The STRINGdatabase in 2011: functional interaction networks of proteins,globally integrated and scored,” Nucleic Acids Research, vol. 39,supplement 1, pp. D561–D568, 2011.

[45] P. Shannon, A. Markiel, O. Ozier et al., “Cytoscape: a softwareenvironment for integrated models of biomolecular interactionnetworks,” Genome Research, vol. 13, no. 11, pp. 2498–2504,2003.

[46] M. J. L. de Hoon, S. Imoto, J. Nolan, and S. Miyano, “Opensource clustering software,” Bioinformatics, vol. 20, no. 9, pp.1453–1454, 2004.

[47] A. J. Saldanha, “Java Treeview-extensible visualization ofmicroarray data,” Bioinformatics, vol. 20, no. 17, pp. 3246–3248,2004.

[48] K. Miyashita, S. Tanakamaru, T. Maitani, and K. Kimura,“Recovery responses of photosynthesis, transpiration, andstomatal conductance in kidney bean following drought stress,”Environmental and Experimental Botany, vol. 53, no. 2, pp. 205–214, 2005.

[49] M. R. B. Siddique, A. Hamid, and M. S. Islam, “Droughtstress effects on water relations of wheat,” Botanical Bulletin ofAcademia Sinica, vol. 41, no. 1, pp. 35–39, 2000.

[50] D. R. Ort, “When there is too much light,” Plant Physiology, vol.125, no. 1, pp. 29–32, 2001.

[51] D. W. Lawlor and G. Cornic, “Photosynthetic carbon assimila-tion and associated metabolism in relation to water deficits inhigher plants,” Plant, Cell & Environment, vol. 25, no. 2, pp. 275–294, 2002.

[52] G. D. Farquhar and T. D. Sharkey, “Stomatal conductance andphotosynthesis,” Annual Review of Plant Physiology, vol. 33, no.1, pp. 317–345, 1982.

[53] A. R. Reddy, K. V. Chaitanya, andM. Vivekanandan, “Drought-induced responses of photosynthesis and antioxidantmetabolism in higher plants,” Journal of Plant Physiology,vol. 161, no. 11, pp. 1189–1202, 2004.

[54] D. W. Lawlor, “Limitation to photosynthesis in water-stressedleaves: stomata vs. Metabolism and the role of ATP,” Annals ofBotany, vol. 89, pp. 871–885, 2002.

[55] H. Medrano, M. A. J. Parry, X. Socias, and D. W. Lawlor, “Longterm water stress inactivates Rubisco in subterranean clover,”Annals of Applied Biology, vol. 131, no. 3, pp. 491–501, 1997.

[56] J. C. V.Vu, R.W.Gesch, L.H.Allen Jr., K. J. Boote, andG. Bowes,“CO2enrichment delays a rapid, drought-induced decrease in

rubisco small subunit transcript abundance,” Journal of PlantPhysiology, vol. 155, no. 1, pp. 139–142, 1999.

[57] B. Moore and J. R. Seeman, “Evidence that 2-carboxyarabinitol1-phosphate binds to ribulose-1, 5-bisphosphate carboxylase invivo,” Plant Physiology, vol. 105, no. 2, pp. 731–737, 1994.

[58] M. M. Chaves, J. P. Maroco, and J. S. Pereira, “Understandingplant responses to drought—from genes to the whole plant,”Functional Plant Biology, vol. 30, no. 3, pp. 239–264, 2003.

[59] M. Havaux and F. Tardy, “Temperature-dependent adjustmentof the thermal stability of photosystem II in vivo: possibleinvolvement of xanthophyll-cycle pigments,” Planta, vol. 198,no. 3, pp. 324–333, 1996.

[60] J. H. Lee, I. H. Kang, K. L. Choi, W.-S. Sim, and J.-K. Kim,“Gene expression of chloroplast translation elongation factor Tuduring maize chloroplast biogenesis,” Journal of Plant Biology,vol. 40, no. 4, pp. 227–233, 1997.

[61] T. Moriarty, R. West, G. Small, D. Rao, and Z. Ristic, “Het-erologous expression of maize chloroplast protein synthesiselongation factor (EF-Tu) enhances Escherichia coli viabilityunder heat stress,” Plant Science, vol. 163, no. 6, pp. 1075–1082,2002.


[62] Z. Ristic, I. Momcilovic, J. Fu, E. Callegari, and B. P. DeRidder,“Chloroplast protein synthesis elongation factor, EF-Tu, reducesthermal aggregation of rubisco activase,” Journal of PlantPhysiology, vol. 164, no. 12, pp. 1564–1571, 2007.

[63] I. Momcilovic and Z. Ristic, “Localization and abundance ofchloroplast protein synthesis elongation factor (EF-Tu) and heatstability of chloroplast stromal proteins inmaize,” Plant Science,vol. 166, no. 1, pp. 81–88, 2004.

[64] D. Rao, I. Momcilovic, S. Kobayashi, E. Callegari, and Z.Ristic, “Chaperone activity of recombinant maize chloroplastprotein synthesis elongation factor, EF-Tu,” European Journal ofBiochemistry, vol. 271, no. 18, pp. 3684–3692, 2004.

[65] M. E. Salvucci, “Association of Rubisco activase with chapero-nin-60𝛽: a possible mechanism for protecting photosynthesisduring heat stress,” Journal of Experimental Botany, vol. 59, no.7, pp. 1923–1933, 2008.

[66] M. Latijnhouwers, X.-M. Xu, and S. G. Møller, “Arabidopsisstromal 70-kDa heat shock proteins are essential for chloroplastdevelopment,” Planta, vol. 232, no. 3, pp. 567–578, 2010.

[67] P.-H. Su and H.-M. Li, “Arabidopsis stromal 70-kD heat shockproteins are essential for plant development and important forthermotolerance of germinating seeds,” Plant Physiology, vol.146, no. 3, pp. 1231–1241, 2008.

[68] M. R. Badger and G. D. Price, “The role of carbonic anhydrasein photosynthesis,” Annual Review of Plant Biology, vol. 45, no.1, pp. 369–392, 1994.

[69] M. H. C. de Carvalho, “Drought stress and reactive oxygenspecies,” Plant Signaling & Behavior, vol. 3, no. 3, pp. 156–165,2008.

[70] S. R. Raisanen, P. Lehenkari, M. Tasanen, P. Rahkila, P. L.Harkonen, and H. K. Vaananen, “Carbonic anhydrase IIIprotects cells from hydrogen peroxide-induced apoptosis,”TheFASEB Journal, vol. 13, no. 3, pp. 513–522, 1999.

[71] N. Tuteja, “Abscisic acid and abiotic stress signaling,” PlantSignaling and Behavior, vol. 2, no. 3, pp. 135–138, 2007.

[72] P. J. Kramer, “Changing concepts regarding plant water rela-tions,” Plant, Cell & Environment, vol. 11, no. 7, pp. 565–568,1988.

[73] Q. Chen and C. D. Silflow, “Isolation and characterizationof glutamine synthetase genes in Chlamydomonas reinhardtii,”Plant Physiology, vol. 112, no. 3, pp. 987–996, 1996.

[74] S. Yan, Z. Tang, W. Su, and W. Sun, “Proteomic analysis of saltstress-responsive proteins in rice root,” Proteomics, vol. 5, no. 1,pp. 235–244, 2005.

[75] R. T. El-Khatib, E. P. Hamerlynck, F. Gallardo, and E. G. Kirby,“Transgenic poplar characterized by ectopic expression of apine cytosolic glutamine synthetase gene exhibits enhancedtolerance to water stress,” Tree Physiology, vol. 24, no. 7, pp. 729–736, 2004.

[76] C. Xu and B.Huang, “Root proteomic responses to heat stress intwo Agrostis grass species contrasting in heat tolerance,” Journalof Experimental Botany, vol. 59, no. 15, pp. 4183–4194, 2008.

[77] S. S. Gill, N. A. Anjum, R. Gill, M.Mahajan, andN. Tuteja, “Abi-otic stress tolerance and sustainable agriculture: a functionalgenomics perspective,” in Elucidation of Abiotic Stress Signalingin Plants, G. K. Pandey, Ed., pp. 439–472, Springer, New York,NY, USA, 2015.

[78] L.-S. P. Tran and K. Mochida, “Identification and prediction ofabiotic stress responsive transcription factors involved in abioticstress signaling in soybean,” Plant Signaling & Behavior, vol. 5,no. 3, pp. 255–257, 2010.

[79] I. Alam, S. A. Sharmin, K.-H.Kim, J. K. Yang,M. S. Choi, andB.-H. Lee, “Proteome analysis of soybean roots subjected to short-term drought stress,” Plant and Soil, vol. 333, no. 1, pp. 491–505,2010.

[80] P. R. Weisz, R. F. Denison, and T. R. Sinclair, “Response todrought stress of nitrogen fixation (acetylene reduction) ratesby field-grown soybeans,” Plant Physiology, vol. 78, no. 3, pp.525–530, 1985.

[81] R. Mittler, “Abiotic stress, the field environment and stresscombination,” Trends in Plant Science, vol. 11, no. 1, pp. 15–19,2006.

[82] H. Vanderschuren, E. Lentz, I. Zainuddin, and W. Gruissem,“Proteomics of model and crop plant species: status, currentlimitations and strategic advances for crop improvement,”Journal of Proteomics, vol. 93, pp. 5–19, 2013.

[83] H. Medrano, J. M. Escalona, J. Bota, J. Gulıas, and J. Flexas,“Regulation of photosynthesis of C

3plants in response to

progressive drought: stomatal conductance as a referenceparameter,” Annals of Botany, vol. 89, pp. 895–905, 2002.

[84] G. Cornic, “Drought stress inhibits photosynthesis by decreas-ing stomatal aperture—not by affecting ATP synthesis,” Trendsin Plant Science, vol. 5, no. 5, pp. 187–188, 2000.

[85] E. S. Tozzi, H. M. Easlon, and J. H. Richards, “Interactive effectsof water, light and heat stress on photosynthesis in Fremontcottonwood,” Plant, Cell & Environment, vol. 36, no. 8, pp. 1423–1434, 2013.

[86] W. Tezara, V. J.Mitchell, S. D. Driscoll, andD.W. Lawlor, “Waterstress inhibits plant photosynthesis by decreasing couplingfactor and ATP,” Nature, vol. 401, no. 6756, pp. 914–917, 1999.

[87] D. Ghosh and J. Xu, “Abiotic stress responses in plant roots: aproteomics perspective,” Frontiers in Plant Science, vol. 5, article6, 13 pages, 2014.

[88] C. Brun, F. Chevenet, D. Martin, J. Wojcik, A. Guenoche, and B.Jacq, “Functional classification of proteins for the prediction ofcellular function from a protein-protein interaction network,”Genome Biology, vol. 5, no. 1, pp. R6.3–R6.13, 2003.

[89] A. Kumar, S. Agarwal, J. A. Heyman et al., “Subcellular localiza-tion of the yeast proteome,” Genes and Development, vol. 16, no.6, pp. 707–719, 2002.

[90] M. A. J. Parry, P. J. Andralojc, S. Khan, P. J. Lea, and A. J. Keys,“Rubisco activity: effects of drought stress,” Annals of Botany,vol. 89, pp. 833–839, 2002.

[91] M. C. Dias andW. Bruggemann, “Limitations of photosynthesisin Phaseolus vulgaris under drought stress: gas exchange,chlorophyll fluorescence and Calvin cycle enzymes,” Photosyn-thetica, vol. 48, no. 1, pp. 96–102, 2010.

[92] H. Cai, Y. Zhou, J. Xiao, X. Li, Q. Zhang, and X. Lian,“Overexpressed glutamine synthetase gene modifies nitrogenmetabolism and abiotic stress responses in rice,” Plant CellReports, vol. 28, no. 3, pp. 527–537, 2009.

[93] J. A. G. Silveira, A. R. B. Melo, R. A. Viegas, and J. T. A. Oliveira,“Salinity-induced effects on nitrogen assimilation related togrowth in cowpea plants,” Environmental and ExperimentalBotany, vol. 46, no. 2, pp. 171–179, 2001.

[94] R. R.Wise, A. J. Olson, S. M. Schrader, and T. D. Sharkey, “Elec-tron transport is the functional limitation of photosynthesis infield-grown Pima cotton plants at high temperature,” Plant, Cell& Environment, vol. 27, no. 6, pp. 717–724, 2004.

[95] R. Zhang and T. D. Sharkey, “Photosynthetic electron transportand proton flux under moderate heat stress,” PhotosynthesisResearch, vol. 100, no. 1, pp. 29–43, 2009.


[96] S. K. Bhadula, T. E. Elthon, J. E. Habben, T. G. Helentjaris, S.Jiao, and Z. Ristic, “Heat-stress induced synthesis of chloroplastprotein synthesis elongation factor (EF-Tu) in a heat-tolerantmaize line,” Planta, vol. 212, no. 3, pp. 359–366, 2001.

[97] M. Haslbeck, T. Franzmann, D. Weinfurtner, and J. Buchner,“Some like it hot: the structure and function of small heat-shockproteins,” Nature Structural and Molecular Biology, vol. 12, no.10, pp. 842–846, 2005.

[98] M. Dalal, D. Tayal, V. Chinnusamy, and K. C. Bansal, “Abioticstress and ABA-inducible Group 4 LEA from Brassica napusplays a key role in salt and drought tolerance,” Journal ofBiotechnology, vol. 139, no. 2, pp. 137–145, 2009.

[99] M. A. Hossain, J.-I. Cho, M. Han et al., “The ABRE-bindingbZIP transcription factor OsABF2 is a positive regulator ofabiotic stress and ABA signaling in rice,” Journal of PlantPhysiology, vol. 167, no. 17, pp. 1512–1520, 2010.

[100] E. McCormack and J. Braam, “Calmodulins and related poten-tial calcium sensors of Arabidopsis,” New Phytologist, vol. 159,no. 3, pp. 585–598, 2003.

[101] M. Fujita, Y. Fujita, Y.Noutoshi et al., “Crosstalk between abioticand biotic stress responses: a current view from the points ofconvergence in the stress signaling networks,” Current Opinionin Plant Biology, vol. 9, no. 4, pp. 436–442, 2006.

[102] S. Komatsu, H.-P. Mock, P. Yang, and B. Svensson, “Applicationof proteomics for improving crop protection/artificial regula-tion,” Frontiers in Plant Science, vol. 4, no. 522, pp. 1–3, 2013.

Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation http://www.hindawi.com

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttp://www.hindawi.com

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research


International Journal of

Microbiology

research article leaf proteome analysis reveals...

Documents